CN114446049B - Traffic flow prediction method, system, terminal and medium based on social value orientation - Google Patents
Traffic flow prediction method, system, terminal and medium based on social value orientation Download PDFInfo
- Publication number
- CN114446049B CN114446049B CN202111637257.4A CN202111637257A CN114446049B CN 114446049 B CN114446049 B CN 114446049B CN 202111637257 A CN202111637257 A CN 202111637257A CN 114446049 B CN114446049 B CN 114446049B
- Authority
- CN
- China
- Prior art keywords
- social value
- track
- vehicle
- target vehicle
- traffic flow
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 62
- 238000004364 calculation method Methods 0.000 claims abstract description 18
- 230000008846 dynamic interplay Effects 0.000 claims abstract description 6
- 230000006399 behavior Effects 0.000 claims description 50
- 239000011159 matrix material Substances 0.000 claims description 20
- 230000001133 acceleration Effects 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 14
- 230000009471 action Effects 0.000 claims description 11
- 238000011156 evaluation Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 description 15
- 238000010586 diagram Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 4
- 230000003993 interaction Effects 0.000 description 4
- 230000011273 social behavior Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 230000004927 fusion Effects 0.000 description 3
- 238000013178 mathematical model Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 230000008092 positive effect Effects 0.000 description 2
- 230000002035 prolonged effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
Landscapes
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention belongs to the technical field of automatic driving, and discloses a traffic flow prediction method, a traffic flow prediction system, a traffic flow prediction terminal and a traffic flow prediction medium based on social value orientation. Dynamic interaction among all vehicle individuals in the traffic flow in the scene is captured by using the game theory, the selfiness and the literacy of the driving behavior of the driving vehicle are quantified by using the social value orientation, and the social value orientation is blended into calculation of traffic flow prediction to predict the driving behavior of the driving vehicle. According to the method, the game theory is used for capturing dynamic interaction among all vehicle individuals in the scene, the parameter of social value orientation is introduced for quantifying the selfiness and the handiness of the driving behavior of the human driver, and the parameter is integrated into the calculation of traffic flow prediction, so that the driving behavior of the human driver can be stably and effectively predicted. The invention quantifies the degree of selfish or literacy of the vehicle driver and can better predict how the vehicle driver will interact and cooperate with others.
Description
Technical Field
The invention belongs to the technical field of automatic driving, and particularly relates to a traffic flow prediction method, a traffic flow prediction system, a traffic flow prediction terminal and a traffic flow prediction medium based on social value orientation.
Background
At present, urban traffic flow has the characteristics of nonlinearity, self-organization, spatiotemporal property, random time variability, periodic similarity and the like, so that the short-time traffic flow prediction technology is complex and difficult.
The main reason for the problems is that human drivers can adopt different driving ideas when facing the same driving scene, drivers with different characters can adopt different driving plans when facing traffic scenes with game characteristics such as parallel lines, left turns, traffic intersections and the like, and even the same driver can continuously change the driving style when facing complex and random time-varying traffic scenes. In summary, it is important to accurately predict traffic flow to effectively control traffic flow and further reduce occurrence probability of bad traffic conditions in the traffic scene at present because vehicles are increasingly complicated.
In view of the above problems, there has been some research and achievements:
the patent CN113240904A relates to a traffic flow prediction method based on feature fusion, which is used for acquiring historical traffic data, performing space correlation analysis, and respectively performing space-time feature extraction by using a graph convolution neural network and a convolution neural network.
The patent CN112991741a provides a traffic flow prediction method and device, in the method, firstly, sensing data of each traffic participant in a target area is obtained through a plurality of intelligent roadside units, then real-time traffic flow data of each traffic participant in the target area in a first preset time period is obtained through data fusion, and predicted traffic flow data of the target traffic participant in a second preset time period is obtained based on a traffic flow prediction model.
The patent CN112863183A relates to a traffic flow data fusion method and system, when detecting traffic flow of a preselected road section, a millimeter wave detection technology and a video image detection technology are adopted at the same time, a display area corresponding to the video image detection technology and a display area corresponding to the millimeter wave detection technology are respectively equally divided into a plurality of first subareas and a plurality of second subareas, so that the fusion speed can be greatly improved, and the integrity of traffic flow information is also improved.
Patent CN113345233a discloses a road traffic flow prediction method and system, which acquires a historical road traffic flow dataset in a selected area for a period of time, and performs preprocessing on the dataset; establishing a traffic flow prediction basic model; and carrying out deep training learning on the traffic flow prediction basic model by utilizing the preprocessed historical road traffic flow data set to obtain optimal parameters, thereby obtaining a traffic flow final prediction model.
Patent CN113326974a proposes a multi-source traffic passenger flow prediction method based on a multi-task hypergraph convolutional network, which relates to the fields of deep learning and the like, in particular to a traffic prediction task oriented to hypergraph representation and graph convolution network. The method is applicable to multi-source heterogeneous traffic data as compared to a single data driven model. The task of rail transit passenger flow prediction is completed, and the prediction accuracy is improved.
The patent CN113240922A discloses a traffic event early warning device based on space-time mutation of traffic flow, and the invention can enhance the prompting performance and simultaneously enable the visibility of a lamp body to be higher by arranging a floating plate; through setting up the display, can be convenient for people to carry out the exact road and select when improving the functionality of traffic lights.
Through the above analysis, the problems and defects existing in the prior art are as follows:
(1) The related patents in the prior art mostly perform some basic data processing based on the acquired data of the prior sensors, or adopt a method based on rule logic or a method based on learning, and the obtained result forms are all represented as mappings from the sensor data to future behaviors, but the basic objects of traffic flow operation are human drivers, and the human drivers have random character characteristics, so that the prediction method in the fixed rule frame form cannot realize stronger robustness in the actual application process, and the accuracy of the predicted driving behavior data is lower.
(2) The driving style of a human driver is influenced by factors such as personal characters, moods and the like, and accurate capturing and prediction are difficult, so that unpredictable driving behaviors are commonly existed in traffic flow, and traffic safety hidden danger is formed.
(3) The vehicle-to-vehicle interactions, i.e., person-to-person interactions, within the traffic stream are random in the consequences that may occur within the gaming scenario, and the observability of the consequences of the process further decreases as the number of participants increases.
(4) At present, in the automatic driving field, the research process of planning decisions has longer periodicity, from information acquisition and processing to actual actions of vehicles, the whole planning decision process covers longer time sequences, in the period, the surrounding vehicle states in the environment where the vehicles are located have stronger random changes, and the track prediction scheme based on the vehicle dynamics model only has instant effectiveness, so that the effectiveness of the whole planning decision process is affected.
The difficulty of solving the problems and the defects is as follows:
(1) The prediction model based on the vehicle dynamics predicts the future motion and the running track of the target vehicle by the speed and the yaw rate of the current vehicle, and is based on the assumption that the driver maintains the existing driving behavior at the current moment. Human drivers often experience more random driving behavior variations due to environmental impact, and therefore the effective time of the basic model prediction results based on vehicle dynamics is limited to 1 to 2s, i.e. the transition time required by the human driver to generate new driving behavior. In the face of more complex scenarios, the long-term traffic flow predictions required contradict the instant effectiveness of existing model-based predictions.
(2) Along with the gradual increase of the number of vehicle nodes in a complex scene, the environment complexity is increased. The relative position relationship, the social relationship and the like among the nodes in the scene have strong random variability, and all possible scene characteristics are difficult to cover by using a common mathematical model based on classification. For the social relationship between the vehicle nodes, it is also difficult to describe efficiently using mathematical models.
(3) The driving style of a human driver plays a decisive role in the formation of future behavior tracks of a vehicle driven by the driver, but the driving style of the human driver is difficult to quantitatively analyze in the actual prediction process, so that the factor cannot be directly and effectively incorporated into the existing mathematical model.
(4) In the course of making a behavior decision using traffic flow prediction data, each decision process is made based on the environmental information that is currently acquired. However, the actual action time of the vehicle has certain hysteresis, and in complex and high-speed traffic scenes (such as highways, traffic intersections and the like), when the vehicle actually performs actions, the scenes are changed greatly, so that the dislocation of the action and behavior decision of the vehicle is caused, and the traffic problem is caused.
The meaning of solving the problems and the defects is as follows:
(1) In the practical application scenario, the data obtained by the prediction requirement of the traffic flow is continuously and effectively available, so that the instant effective prediction method is not applicable. After the problems are solved, more global information can be obtained, and a more abundant and effective data basis is provided for subsequent behavior decisions.
(2) The social relations among the vehicle nodes are quantized and incorporated into the input of the prediction algorithm, so that the model is more comprehensive in understanding of the current scene, the prediction model is more complete, and the accuracy of traffic flow prediction is effectively improved.
(3) The hysteresis of the vehicle action is considered in the calculation model, so that the planning decision process based on traffic flow prediction can be carried out aiming at the moment (namely the future moment) of the actual action of the vehicle, thereby ensuring the corresponding relation of data acquisition and bottom control and effectively avoiding traffic problems caused by delay.
Disclosure of Invention
In order to overcome the problems in the related art, the disclosed embodiments of the invention provide a traffic flow prediction method, a traffic flow prediction system, a traffic flow prediction terminal and a traffic flow prediction medium based on social value orientation.
The technical scheme is as follows: a traffic flow prediction method based on social value orientation uses game theory to capture dynamic interaction among all traffic flow vehicle individuals in a scene, utilizes the social value orientation to quantify the selfiness and the literacy of driving behavior of a driving vehicle, and blends the social value orientation into calculation of global traffic flow prediction to carry out more accurate global prediction on the driving behavior of the driving vehicle.
In one embodiment, the traffic flow prediction method based on social value orientation comprises the following steps:
step one, evaluating driving behaviors of a driving vehicle by using social value orientation;
step two, measuring and evaluating the social value orientation in real time;
and thirdly, predicting the target vehicle based on the social value orientation.
In one embodiment, the step of evaluating driving behavior of the driving vehicle using the social value orientation includes:
integrating a utility function g (-) of social value orientation into a non-cooperative dynamic game, and modeling a vehicle driver to obtain a maximized utility value; the weighted value of the utility function g (-) is obtained by calculating the social value orientation, and the utility function g (-) is as follows:
wherein r is 1 And r 2 Respectively the utility of itself and other vehicle utilities,is the social value orientation value of the target vehicle.
In an embodiment, the obtained maximized utility value includes:
ritual sense: maximizing game-to-cube utility without combining the vehicle driver's own results, corresponding to
Sociality is a group: the intention of the behavior of the vehicle driver is to maximize the utility of the entire population, corresponding to
Lithosense: maximizing the utility of the vehicle driver, not combining the utility of the pair cubes, corresponds to
Contentment: maximizing the utility ratio of the vehicle driver to the cube, corresponding to
In one embodiment, the step of real-time measurement and evaluation of the second social value orientation includes: step 1, classifying expected tracks formed by different social value orientations, wherein the classification of the expected tracks can be different according to the change of actual traffic scenes:
in a multi-lane same-direction straight-driving traffic scene, the expected track can be divided into state keeping, accelerating preemption lanes, decelerating and avoiding, left lane changing, right lane changing and the like according to different social value orientations;
in a multi-lane opposite direction straight driving traffic scene, the expected track can be basically the same as the same direction straight driving, and can be divided into state keeping, acceleration preempting lanes, deceleration avoiding, left lane changing, right lane changing, left same lane fine adjustment, right same lane fine adjustment and the like according to different social value orientations;
in a single-lane same-direction straight-driving traffic scene, the expected track can be divided into state maintenance, fine adjustment of occupied road, right fine adjustment of yielding lane, acceleration straight driving, deceleration and the like according to different social value orientations;
in a single-lane opposite-direction straight-driving traffic scene, the expected track can be divided into state maintenance, lane occupation fine adjustment, lane yielding of right fine adjustment, lane preemption of left fine adjustment, parking avoidance by side and the like according to different social value orientations;
in a traffic intersection scene, the planned intersection behaviors of the vehicle, such as straight, right turn, left turn, head drop and the like, are known according to the vehicle signal lamp. According to different social value orientations, the expected track can be divided into the steps of keeping the original planning behavior, parking waiting, suspending the original planning behavior, canceling the original planning behavior, selecting a new driving behavior and the like;
in summary, in different traffic scenes, the predicted tracks can be classified according to different social value orientations;
step 2, after classifying to form a predicted track data set, comparing the predicted track data set with an actual track, wherein the probability and distribution of candidate social value orientation values are calculated by calculating the distance between the predicted track and the actual track based on one condition, and the social value orientation is measured and evaluated in real time, and the specific implementation process is as follows:
performing deviation calculation on different predicted tracks and actual tracks of the vehicle obtained by observation, wherein the deviation is represented as an expected value of Euclidean distance between corresponding track points, and a calculation formula is as follows:
wherein Δ is the result of the deviation calculation, (x) i ,y i ) In order to predict the plane coordinates of the trajectory,plane coordinates of the actual trajectory are obtained for observation.
Step 3, selecting the social value orientation corresponding to the most-matched predicted track as the social value orientation judgment value of the target vehicle according to the deviation value sequence of different predicted tracks for subsequent further track prediction;
the formula is:
k＝argminΔ k
k represents the social value orientation value, namely the expected value.
In one embodiment, the predicting the target vehicle based on the social value orientation includes:
(1) Setting the current time as T 0 Setting verification time delta T as dynamic update observation time length for judging SVO value of target vehicle based on T 0 - Δt time to current time T 0 And observing the obtained actual track information of the target vehicle to judge the current optimal SVO predicted value of the target vehicle, wherein the predicted value is a real-time updated value and is used for inputting a predicted model of a subsequent future track.
(2) Capturing a target vehicle information sequence matrix based on a sensor, wherein the target vehicle information sequence matrix comprises longitudinal vehicle speed, longitudinal acceleration, yaw rate and yaw rate acceleration, and the specific matrix has the following expression form:
wherein in is the information sequence matrix of the target vehicle, v i A, for the longitudinal speed of the target vehicle at the corresponding moment i For the corresponding moment of longitudinal acceleration of the target vehicle omega i Yaw rate, alpha, for the target vehicle at the corresponding instant i Yaw angular acceleration at the corresponding moment of the target vehicle;
the information sequence matrix in of the target vehicle is input into the existing vehicle dynamics-based model to generate a predicted track thereof, so as to improve the prediction breadth and ensure that a predicted result contains a real track, and the generated predicted track is expressed in a sequence form, namely, a distribution is formed for future tracks of the target vehicle:
wherein TR is a track distribution matrix,for the target vehicle position coordinates +.>For the moment of position coordinate correspondence, then +.>Namely, the track point of the target vehicle;
(3) Comparing the distribution matrix TR of the predicted track with behavior classification based on the current social value orientation to obtain a track TR with the most consistent space-time position k As a final predicted trajectory, for prediction of current traffic flow:
k＝argminδ k
wherein delta k For predicting the deviation of the track from the actual trackThe difference is expressed as an expected value of euclidean distance in the spatio-temporal range.
In an embodiment, the controlling the traffic flow based on the predicted track of the target vehicle further includes:
and carrying out overall optimal planning control on all vehicles in the current scene based on the obtained traffic flow prediction data, obtaining the next optimal walking solution of each vehicle, and issuing the operation result to a bicycle to carry out control of expected actions.
Another object of the present invention is to provide a traffic flow prediction system based on social value orientation, comprising:
the driving behavior evaluation module is used for evaluating the driving behavior of the driving vehicle by using the social value orientation;
the social value orientation evaluation module is used for measuring and evaluating the social value orientation in real time;
and the target vehicle prediction module is used for predicting the target vehicle based on the social value orientation.
And the traffic flow control module is used for controlling traffic flow based on the predicted track of the target vehicle.
It is another object of the present invention to provide a storage medium for receiving user input, the stored computer program causing an electronic device to perform the steps of: capturing dynamic interaction among vehicle individuals in all traffic flows in a scene by using a game theory, quantifying the selfiness and the handiness of driving behavior of a driving vehicle by using social value orientation, and integrating the social value orientation into calculation of traffic flow prediction to predict the driving behavior of the driving vehicle;
and carrying out overall optimal planning control on all vehicles in the current scene based on the obtained traffic flow prediction data, obtaining the next optimal walking solution of each vehicle, and issuing the operation result to a bicycle to carry out control of expected actions.
Another object of the present invention is to provide an information data processing terminal including a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to execute the social value orientation-based traffic flow prediction method.
By combining all the technical schemes, the invention has the advantages and positive effects that:
the invention provides a method based on social value orientation to effectively predict traffic flow, introduces a social psychology research method, captures dynamic interaction among all vehicle individuals in a scene by using a game theory, further introduces a parameter of social value orientation (SVO Social Value Orientation) for quantifying the selfiness and literacy of the driving behavior of a human driver, and integrates the parameter into the calculation of traffic flow prediction, thereby helping to stably and effectively predict the driving behavior of the human driver.
The invention integrates the socioeconomic tool into the traffic flow prediction to quantify and predict the social behavior of other drivers, wherein one key component is the Social Value Orientation (SVO), which quantifies the selfish or literacy degree of the vehicle driver in real time, and can better predict how the vehicle driver will interact and cooperate with other people, thereby improving the accuracy of traffic flow prediction.
Advantages of the present invention compared to the prior art further include:
integrating a socio-psychological tool into autonomous vehicle decisions to quantify and predict social behavior of other drivers, one of the key components being Social Value Orientation (SVO);
the SVO value is used for carrying out effective mathematical description on the social relationship between vehicle nodes in the traffic scene, so that the data input of a track prediction model is supplemented, and the global property of track prediction is improved;
the historical track sequence of the target vehicle is observed and updated in real time, the SVO value of the target vehicle is calculated and updated in real time, and the psychological fluctuation of the driver caused by random and changeable traffic conditions in a complex traffic environment can be effectively captured and adapted;
compared with the existing track prediction model based on vehicle dynamics and learning, the method can obtain the distribution of the track prediction values in advance so as to ensure the prediction breadth, namely, ensure that the actual track is in the set of the prediction results, and then screen the actual track through SVO values, so that the calculation fault tolerance of the model can be effectively improved;
the interpretation of future driving behaviors of the target vehicle is enhanced by integrating the concept of the social value orientation SVO in the model, and the cognition of the model to the environment is improved so as to effectively control;
the integration of SVO can explain the social relationship among vehicle nodes, and provides a data base for realizing larger social utility (rewards) by utilizing the characteristics subsequently;
according to the invention, the validity duration of the prediction model is prolonged by considering the social information of the human driver, the global performance of the prediction model is improved, and the method has outstanding compensation and correction capabilities for inherent delay caused by a hardware system and calculation cost in the subsequent planning decision process;
the invention utilizes a statistical related mathematical tool in the data processing part, and improves the robustness of model calculation in a probability distribution mode;
by utilizing the development characteristics of big data, the SVO judgment model can be trained by means of the data set, so that a faster SVO solving model is obtained.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure of the invention as claimed.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure.
Fig. 1 is a flow chart of a traffic flow prediction and control method based on social value orientation provided by an embodiment of the invention.
FIG. 2 is a diagram showing a comparison of a method for incorporating SVO prediction and a prior art method according to an embodiment of the present invention.
FIG. 3 is a schematic diagram of a Social Value Orientation (SVO) value provided by an embodiment of the invention.
FIG. 4 is a schematic diagram of the effect of SVO on vehicle behavior according to an embodiment of the present invention.
Fig. 5 is a diagram of an SVO estimation process according to an embodiment of the present invention, where fig. 5 (a) is a graph of the SVO estimation process, and fig. 5 (b) is a histogram of the SVO estimation process.
Fig. 6 is a statistical diagram of SVO distribution of a vehicle according to an embodiment of the present invention.
Detailed Description
In order that the above objects, features and advantages of the invention will be readily understood, a more particular description of the invention will be rendered by reference to the appended drawings. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. The invention may be embodied in many other forms than described herein and similarly modified by those skilled in the art without departing from the spirit or scope of the invention, which is therefore not limited to the specific embodiments disclosed below.
Integrating a socio-psychological tool into autonomous vehicle decisions to quantify and predict social behavior of other drivers, one of the key components being Social Value Orientation (SVO);
the traffic flow prediction and control method based on social value orientation provided by the embodiment of the invention comprises the following steps: by estimating the SVO of the vehicles in the traffic flow in real time, more accurate track prediction is realized; the social relationship among the vehicle nodes in the traffic scene is effectively described mathematically through the SVO value, so that the data input of the track prediction model is supplemented, and the global property of track prediction is improved; the interpretation of future driving behaviors of the target vehicle is enhanced by integrating the concept of social value orientation SVO in the model; the validity duration of the prediction model is prolonged by considering the social information of the human driver, and the global performance of the prediction model is improved; social Value Orientation (SVO) improves the ability of fine cooperation between vehicles, thereby achieving greater social utility (rewards); the track is predicted in a probability distribution mode, so that the calculation robustness of the model is improved, and the noise influence is effectively controlled; by utilizing the development characteristics of big data, the SVO solver is trained through a data set, so that a solution model with stronger coverage is obtained.
The invention also provides a traffic flow prediction system based on social value orientation, which comprises:
the driving behavior evaluation module is used for evaluating the driving behavior of the driving vehicle by using the social value orientation;
the social value orientation evaluation module is used for measuring and evaluating the social value orientation in real time;
and the target vehicle prediction module is used for predicting the target vehicle based on the social value orientation.
And the traffic flow control module is used for controlling traffic flow based on the predicted track of the target vehicle.
The technical scheme of the invention is further described below with reference to specific embodiments.
Examples
The present invention models interactions between vehicle drivers as a best response game in which each agent negotiates to maximize its own utility. The invention calculates the social value orientation (SVO Social Value Orientation) of the target vehicles on the basis of capturing the historical tracks of the target vehicles, and further provides a method for predicting interaction of multiple intelligent agents on line. Namely, the traffic flow prediction and control method based on social value orientation, as shown in fig. 1, the specific operation flow is divided into the following steps:
s101, evaluating driving behavior of a human driver using Social Value Orientation (SVO):
the invention integrates Social Value Orientation (SVO) into a non-cooperative dynamic game, and models a vehicle driver so as to quantitatively evaluate the social behavior of the vehicle driver. In order to integrate the Social Value Orientation (SVO) into the formula for efficient quantification, the present invention defines a utility (rewards) function g (·) that combines the utility (rewards) values of the host vehicle and other vehicles around, the weighted values being determined computationally from the Social Value Orientation (SVO), for example, in a two-player game process, the function can be defined as:
wherein r is 1 And r 2 Respectively, self utility (rewards) and other vehicle utility (rewards),for the Social Value Orientation (SVO) value of the target vehicle, the present invention can list the following common SVO values and their corresponding driving styles according to the function:
ritual sense: maximizing game-to-cube utility (rewards) by vehicle drivers without considering their own results, corresponding to
Sociality is a group: the intention of the behavior of the vehicle driver is to maximize the utility (rewards) of the whole population, corresponding to
Lithosense: maximizing the utility (rewards) of the vehicle driver himself, irrespective of the utility (rewards) to the cube, corresponding to
Contentment: maximizing the utility (rewarding) ratio of the vehicle driver himself to the cube, corresponding to
S102, measuring and evaluating Social Value Orientation (SVO) in real time:
the present invention can predict future driving behavior of a vehicle by observing a historical track of the vehicle and estimating a Social Value Orientation (SVO) value, wherein the SVO value has a decisive role for track prediction. In the present invention, FIG. 2 is a diagram providing a comparison of a method of incorporating SVO prediction with prior art methods. FIG. 3 is a schematic diagram of the Social Value Orientation (SVO) values of the present invention.
As shown in FIG. 4, social Value Orientation (SVO) of the parent society generates braking trajectory predictions, while Lignosis SVO generates non-braking trajectory predictions.
As shown in fig. 5 (a) and fig. 5 (b) are graphs of SVO estimation process), the expected trajectories formed by different Social Value Orientations (SVOs) are calculated respectively, and compared with the actual trajectories obtained by observation, and the candidate Social Value Orientations (SVOs) are calculated by calculating the distances between the predicted trajectories and the actual trajectories, so that the Social Value Orientations (SVOs) are measured and evaluated in real time.
S103, predicting the target vehicle based on Social Value Orientation (SVO):
based on a target vehicle information matrix captured by a sensor, which comprises a longitudinal vehicle speed, a longitudinal acceleration, a yaw rate acceleration and the like, a distribution is formed on future tracks of the target vehicle, corresponding points are captured in Social Value Orientation (SVO) polar coordinates according to the predicted tracks, so that tracks which are most in line with the current Social Value Orientation (SVO) value are obtained as final predicted tracks, and prediction of current traffic flow is realized.
S104, controlling traffic flow based on the predicted track of the target vehicle:
and carrying out overall optimal planning control on all vehicles in the current scene based on the acquired more accurate traffic flow prediction data, so as to obtain the next optimal walking solution of each vehicle, and sending the operation result to a bicycle to realize the expected action.
In step S104, based on the obtained more accurate traffic flow prediction data, dynamic map information (specifically expressed in the form of image video) of the vehicle node is formed, so as to visualize the traffic flow calculation result and send the traffic flow calculation result to the vehicle node terminal through the communication module, which can be specifically expressed as: for vehicles with higher intelligent networking degree, the information can be directly issued to the bicycle through the cloud terminal; for vehicles with low intelligent networking degree, information can be transmitted to a bicycle through a mobile terminal (intelligent mobile phone software, an electronic map and the like) held by a human driver; therefore, more global and stronger environmental information is provided for the vehicle nodes, an internal decision system (the unmanned vehicle is embodied as a decision program, and the manned vehicle is embodied as a brain of a human driver) is assisted to form a better driving strategy, and therefore traffic flow is controlled and optimized.
The positive effects of the present invention are further described below in connection with experimental data.
In order to verify the method provided by the invention, the capability of the algorithm for predicting the merging lane action target vehicle track of the expressway entrance ramp is tested in the NGSIM data set. By utilizing the track data of the related vehicle provided by the data set, the SVO-based prediction method provided by the invention is compared with a basic prediction method based on a vehicle kinematic model, and the result is as follows:
| prediction | Basic algorithm | SVO1 | SVO2 | SVO3 |
| SVO-based | —— | Liji sense | Static SVO | Dynamic SVO |
| Track mean square error | 1.0 | 0.947 | 0.821 | 0.753 |
The present invention has found that a prediction method incorporating SVO generally reduces the predicted error value compared to a kinematic model-based algorithm, wherein the target vehicle SVO is always set toThe method reduces the error by 5%, reduces the error by 18% based on the current environment judgment using the static SVO value, and reduces the error by 25% based on the real-time environment updating SVO value.
According to the data test result, the SVO values of the vehicles with the merged lane behaviors and the vehicles keeping the original lanes are counted, as shown in fig. 6, which is a vehicle SVO distribution statistical chart provided by the invention (a thick line frame represents the vehicles keeping the lanes and a thin line frame represents the vehicles with the merged lanes).
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any adaptations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It is to be understood that the present disclosure is not limited to the precise arrangements and instrumentalities shown in the drawings, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the present disclosure should be limited by the attached claims.
Claims (7)
1. The traffic flow prediction method based on the social value orientation is characterized in that the traffic flow prediction method based on the social value orientation captures dynamic interactions among vehicle individuals in all traffic flows in a scene by using a game theory, utilizes the social value orientation to quantify the selfiness and the literacy of driving behaviors of driving vehicles, and integrates the social value orientation into calculation of traffic flow prediction to predict the driving behaviors of the driving vehicles;
the traffic flow prediction method based on social value orientation comprises the following steps:
firstly, measuring and evaluating the orientation of social value in real time;
step two, evaluating the driving behavior of the driving vehicle by using the social value orientation;
predicting the target vehicle based on the social value orientation;
the step of real-time measurement and evaluation of the social value orientation comprises the following steps:
step 1, classifying expected tracks formed by different social value orientations, wherein the classification of the expected tracks is different according to the change of actual traffic scenes:
in a multi-lane same-direction straight-driving traffic scene, the expected track can be divided into state keeping, accelerating preemption lanes, decelerating and avoiding, left lane changing and right lane changing according to different social value orientations;
in a multi-lane opposite direction straight driving traffic scene, the method is basically the same as the same direction straight driving, and the expected track can be divided into state keeping, acceleration preempting lanes, deceleration avoiding, left lane changing, right lane changing, left same lane fine adjustment and right same lane fine adjustment according to different social value orientations;
in a single-lane same-direction straight-driving traffic scene, the expected track can be divided into state keeping, track occupation fine adjustment, lane yielding right fine adjustment, straight driving acceleration and deceleration according to different social value orientations;
in a single-lane opposite-direction straight-driving traffic scene, the expected track can be divided into state keeping, lane occupation fine adjustment, lane yielding of right fine adjustment, lane preemption of left fine adjustment and parking avoidance by side according to different social value orientations;
within a traffic intersection scenario, the planned intersection behavior of a vehicle is known from vehicle signal lights, including: straight, right turn, left turn, turn around;
according to different social value orientations, the expected track can be divided into the steps of keeping the original planning behavior, parking waiting, suspending the original planning behavior, canceling the original planning behavior and selecting a new driving behavior;
step 2, after classifying to form an expected track data set, comparing with an actual track, calculating the possibility and distribution of candidate social value orientation values by calculating the distance between the expected track and the actual track, and measuring and evaluating the social value orientation in real time, wherein the method specifically comprises the following steps:
performing deviation calculation on different expected tracks and actual tracks of the vehicle obtained by observation, wherein the deviation is represented as an expected value of Euclidean distance between corresponding track points, and a calculation formula is as follows:
wherein Δ is the result of the deviation calculation, (x) i ,y i ) For the plane coordinates of the intended trajectory,obtaining plane coordinates of an actual track for observation;
step 3, selecting the social value orientation corresponding to the most-conforming expected track as the social value orientation judgment value of the target vehicle according to the deviation value sequence of different expected tracks, and using the social value orientation judgment value for the subsequent further track prediction;
the formula is:
representing social valueValue orientation values;
the step two of evaluating the driving behavior of the driving vehicle by using the social value orientation comprises the following steps:
defining a utility function of the social value orientation as a weighted sum of the self utility function and other vehicle utility functions and a trigonometric function of the social value orientation value of the target vehicle;
the weight coefficients corresponding to the self utility and other vehicle utilities are respectively a cosine trigonometric function and a sine trigonometric function;
the independent variables corresponding to the sine function and the cosine function are social value orientation values of the target vehicle;
the step three of predicting the target vehicle based on the social value orientation comprises the following steps:
(1) Setting the current time as T 0 Setting verification time delta T as dynamic update observation time length for judging social value orientation SVO value of target vehicle based on T 0 - Δt time to current time T 0 Observing the obtained actual track information of the target vehicle to judge the current social value orientation SVO predicted value of the target vehicle, wherein the predicted value is a real-time updated value and is used for inputting a predicted model of a subsequent future track;
(2) Capturing a target vehicle information sequence matrix based on a sensor, wherein the target vehicle information sequence matrix comprises longitudinal vehicle speed, longitudinal acceleration, yaw rate and yaw rate acceleration; inputting an information sequence matrix in of a target vehicle into an existing model based on vehicle dynamics to generate a target vehicle predicted track, wherein the generated predicted track is the distribution formed on the future track of the target vehicle, and a distribution matrix TR of the predicted track is formed;
(3) Comparing the distribution matrix TR of the predicted track with behavior classification based on the current social value orientation to obtain a track TR with the most consistent space-time position k As a final expected trajectory for prediction of current traffic flow.
2. The traffic flow prediction method based on social value orientation according to claim 1, wherein the obtained maximized utility value includes:
ritual sense: driving without combining with vehicleUnder the condition of self-result of the driver, the utility of the game to the cube is maximized, corresponding to
Sociality is a group: the intention of the behavior of the vehicle driver is to maximize the utility of the entire population, corresponding to
Lithosense: maximizing the utility of the vehicle driver, not combining the utility of the pair cubes, corresponds to
Contentment: maximizing the utility ratio of the vehicle driver to the cube, corresponding to
3. The traffic flow prediction method based on social value orientation according to claim 1, wherein the step three of predicting the target vehicle based on the social value orientation comprises:
(1) Setting the current time as T 0 Setting verification time delta T as dynamic update observation time length for judging social value orientation SVO value of target vehicle based on T 0 - Δt time to current time T 0 Observing the obtained actual track information of the target vehicle to judge the current social value orientation SVO predicted value of the target vehicle, wherein the predicted value is a real-time updated value and is used for inputting a predicted model of a subsequent future track;
(2) Capturing a target vehicle information sequence matrix based on a sensor, wherein the target vehicle information sequence matrix comprises longitudinal vehicle speed, longitudinal acceleration, yaw rate and yaw rate acceleration, and the specific matrix has the following expression form:
wherein in is an information sequence matrix of the target vehicle, v is a longitudinal vehicle speed at the corresponding moment of the target vehicle, a is a longitudinal acceleration at the corresponding moment of the target vehicle, ω is a yaw rate at the corresponding moment of the target vehicle, and α is a yaw rate at the corresponding moment of the target vehicle;
inputting an information sequence matrix in of a target vehicle into an existing vehicle dynamics-based model to generate a target vehicle predicted track, wherein the generated predicted track is a distribution formed on a future track of the target vehicle:
wherein TR is a track distribution matrix,for the target vehicle position coordinates +.>For the moment corresponding to the position coordinates, thenThe track points are target vehicle track points;
(3) Comparing the distribution matrix TR of the predicted track with behavior classification based on the current social value orientation to obtain a track TR with the most consistent space-time position k As a final predicted trajectory, for prediction of current traffic flow:
k＝argminδ k
wherein delta k The expected value of euclidean distance in the space-time range is expressed as the deviation of the predicted trajectory from the actual trajectory.
4. The traffic flow prediction method based on social value orientation according to claim 1, wherein the controlling the traffic flow based on the expected track of the target vehicle further comprises:
and carrying out overall planning control on all vehicles in the current scene based on the obtained traffic flow prediction data, obtaining next walking data of each vehicle, and issuing an operation result to a bicycle to carry out control of expected actions.
5. A social value orientation-based traffic flow prediction system that implements the social value orientation-based traffic flow prediction method of any one of claims 1 to 4, characterized in that the social value orientation-based traffic flow prediction system comprises:
the driving behavior evaluation module is used for evaluating the driving behavior of the driving vehicle by using the social value orientation;
the social value orientation evaluation module is used for measuring and evaluating the social value orientation in real time;
the target vehicle prediction module is used for predicting the target vehicle based on the social value orientation;
and the traffic flow control module is used for controlling traffic flow based on the expected track of the target vehicle.
6. A storage medium receiving a user input program, the stored computer program causing an electronic device to perform the steps of the social value orientation based traffic flow prediction method according to any one of claims 1-4.
7. An information data processing terminal, characterized in that the information data processing terminal comprises a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to execute the traffic flow prediction method based on social value orientation according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111637257.4A CN114446049B (en) | 2021-12-29 | 2021-12-29 | Traffic flow prediction method, system, terminal and medium based on social value orientation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111637257.4A CN114446049B (en) | 2021-12-29 | 2021-12-29 | Traffic flow prediction method, system, terminal and medium based on social value orientation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114446049A CN114446049A (en) | 2022-05-06 |
| CN114446049B true CN114446049B (en) | 2024-03-08 |
Family
ID=81365071
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111637257.4A Active CN114446049B (en) | 2021-12-29 | 2021-12-29 | Traffic flow prediction method, system, terminal and medium based on social value orientation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114446049B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115050183B (en) * | 2022-06-09 | 2023-07-25 | 上海人工智能创新中心 | Method for generating simulated traffic flow |
| CN115221971B (en) * | 2022-07-28 | 2024-06-14 | 上海人工智能创新中心 | Trajectory prediction method based on heterogeneous graph |
| CN115909749B (en) * | 2023-01-09 | 2023-07-14 | 广州通达汽车电气股份有限公司 | Vehicle running road risk early warning method, device, equipment and storage medium |
| CN117275240B (en) * | 2023-11-21 | 2024-02-20 | 之江实验室 | Traffic signal reinforcement learning control method and device considering multiple types of driving styles |
| CN117709602B (en) * | 2024-02-05 | 2024-05-17 | 吉林大学 | Urban intelligent vehicle personification decision-making method based on social value orientation |
| CN118457609B (en) * | 2024-07-10 | 2024-09-03 | 吉林大学 | Method for calculating social value orientation of driver in driving interaction scene |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101756705A (en) * | 2008-11-14 | 2010-06-30 | 北京宣爱智能模拟技术有限公司 | System and method for testing driving accident proneness |
| CN104331953A (en) * | 2014-10-29 | 2015-02-04 | 云南大学 | Car behavior data identification and management method based on internet of things |
| CN112116100A (en) * | 2020-09-08 | 2020-12-22 | 南京航空航天大学 | Game theory decision method considering driver types |
| CN112581832A (en) * | 2020-12-14 | 2021-03-30 | 公安部交通管理科学研究所 | Learning style-based critical driving risk evaluation intervention method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11884302B2 (en) * | 2019-11-15 | 2024-01-30 | Massachusetts Institute Of Technology | Social behavior for autonomous vehicles |
-
2021
- 2021-12-29 CN CN202111637257.4A patent/CN114446049B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101756705A (en) * | 2008-11-14 | 2010-06-30 | 北京宣爱智能模拟技术有限公司 | System and method for testing driving accident proneness |
| CN104331953A (en) * | 2014-10-29 | 2015-02-04 | 云南大学 | Car behavior data identification and management method based on internet of things |
| CN112116100A (en) * | 2020-09-08 | 2020-12-22 | 南京航空航天大学 | Game theory decision method considering driver types |
| CN112581832A (en) * | 2020-12-14 | 2021-03-30 | 公安部交通管理科学研究所 | Learning style-based critical driving risk evaluation intervention method and system |
Non-Patent Citations (2)
| Title |
|---|
| 交通方式选择的非集计模型及其应用;贾洪飞等;《吉林大学学报(工学版)》;20071115(第06期);第66-71页 * |
| 结合车道线检测的智能车辆位姿估计方法;李琳辉等;《科学技术与工程》;20200728(第21期);第387-392页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114446049A (en) | 2022-05-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114446049B (en) | Traffic flow prediction method, system, terminal and medium based on social value orientation | |
| US11726477B2 (en) | Methods and systems for trajectory forecasting with recurrent neural networks using inertial behavioral rollout | |
| CN109774724A (en) | The scene of exploitation & assessment for autonomous driving system generates and the method and apparatus of parameter scanning | |
| Chong et al. | A rule-based neural network approach to model driver naturalistic behavior in traffic | |
| EP3035314B1 (en) | A traffic data fusion system and the related method for providing a traffic state for a network of roads | |
| US11919545B2 (en) | Scenario identification for validation and training of machine learning based models for autonomous vehicles | |
| CN114077541A (en) | Method and system for validating automatic control software for an autonomous vehicle | |
| CN112734808B (en) | Trajectory prediction method for vulnerable road users in vehicle driving environment | |
| CN105513354A (en) | Video-based urban road traffic jam detecting system | |
| CN109272745A (en) | A kind of track of vehicle prediction technique based on deep neural network | |
| CN112258841A (en) | Intelligent vehicle risk assessment method based on vehicle track prediction | |
| CN114299607A (en) | Human-vehicle collision risk degree analysis method based on automatic driving of vehicle | |
| CN109782751A (en) | Method and apparatus for autonomous system performance and benchmark test | |
| Miani et al. | Young drivers’ pedestrian anti-collision braking operation data modelling for ADAS development | |
| Liu et al. | AdaBoost-Bagging deep inverse reinforcement learning for autonomous taxi cruising route and speed planning | |
| Li et al. | A probabilistic framework for estimating the risk of pedestrian-vehicle conflicts at intersections | |
| CN116823572B (en) | Population flow data acquisition method and device and computer readable storage medium | |
| Wen et al. | Modeling human driver behaviors when following autonomous vehicles: An inverse reinforcement learning approach | |
| Sun et al. | Medium-fidelity evaluation and modeling for perception systems of intelligent and connected vehicles | |
| CN113361866A (en) | Two-dimensional simulation method for microscopic traffic flow of non-motor vehicle | |
| Ning | Prediction and detection of urban trajectory using data mining and deep neural network | |
| CN116013093B (en) | Collaborative evacuation simulation method and device based on video | |
| CN116989818B (en) | Track generation method and device, electronic equipment and readable storage medium | |
| Luo et al. | A Driving Intention Prediction Method for Mixed Traffic Scenarios | |
| Nasernejad | Modeling pedestrian behavior in pedestrian-vehicle near misses using inverse reinforcement learning algorithms |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |